Measuring method and application of oil cut rate and total flow rate of oil well based on dense neural network

Abstract

A method for measuring oil content and total flow rate of an oil well based on a dense neural network and its application, including: using four-sector conductance sensors to collect fluid data of different working conditions in the wellhead oil and gas pipeline, obtaining tags, and uploading them to the host computer for storage; constructing data Specifically, after preprocessing the fluid data, use overlapping sliding windows to obtain samples from the fluid data, add corresponding labels to the samples, and create a training set, verification set, and test set; construct a dense neural network model; according to Different training stages use different labels for training. After all training stages are completed, the integrated dense neural network model is used for parameter measurement; the dense neural network model is optimized, trained, and parameter adjusted to obtain the optimal dense neural network model at different stages of the current architecture. The neural network model is combined to complete parameter measurement. The invention can accurately measure the oil content and the total flow velocity of the oil-water two-phase flow in the wellhead environment in real time.

Description

technical field [0001] The invention relates to a method for measuring multi-parameters of oil content and total flow velocity of an oil well. In particular, it relates to a method for measuring oil content and total flow rate of an oil well based on a dense neural network and its application. Background technique [0002] Oil-water two-phase flow exists widely in industrial fields such as oil and gas exploitation, transportation, and processing. Under the oil-water two-phase flow system, due to the difference in density, viscosity and other properties of the two-phase medium, their distribution in the pipeline changes in both space and time, forming a specific flow shape, that is, a flow pattern. With the change of the flow pattern in the two-phase flow system, the overall flow information is also constantly changing, resulting in drastic changes in the key flow parameters (phase flow rate, phase holdup, etc.), and the related flow parameters are also in strong fluctuation...

AI Technical Summary

Problems solved by technology

Traditional annular conductance sensors and double helical capacitive sensors, etc., only contain single-channel fluid information, and cannot accurately reflect the flow structure

Images